| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Epidemiology, University of North Carolina School of Public Health, and the Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina.
Address reprint requests to: Michael J. McMahon, MD, MPH University of North Carolina School of Medicine Department of Obstetrics and Gynecology 214 MacNider Building Campus Box #7516 Chapel Hill, NC 27599-7516 E-mail: mcmahon{at}med.unc.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Using information from birth certificates of live singleton births in North Carolina from 1990 to 1997 (n = 807,759), we evaluated the frequency of intrapartum hemorrhage and its association with maternal race. Logistic regression models were used to evaluate the risk of any intrapartum hemorrhage, placental abruption, placenta previa, and unspecified hemorrhage in each racial group, adjusted for other risk factors.
Results: Black women had the highest rates of any hemorrhage (1.52% black, 1.47% white, 1.33% other race, P = .006) and placental abruption (0.79% black, 0.68% white, 0.56% other race, P = .001) but had lower rates of unspecified hemorrhage (0.37% black, 0.42% white, 0.42% other race, P = .001). Race was not associated with placenta previa. Maternal race remained associated with intrapartum hemorrhage after multivariable analysis, but the direction of the association was reversed. Black women were less likely to have any intrapartum hemorrhage (odds ratio [OR] 0.80, 95% confidence interval [CI] 0.77, 0.85), placental abruption (OR 0.76, 95% CI 0.70, 0.82), placenta previa (OR 0.89, 95% CI 0.81, 0.98), or other unspecified hemorrhage (OR 0.84, 95% CI 0.76, 0.92) compared with white women. Women of other minority races were at lower risk for placental abruption (OR 0.76, 95% CI 0.67, 0.87) but were comparable to white women for risk of placenta previa (OR 1.06, 95% CI 0.91, 1.24) and other unspecified hemorrhage (OR 1.02, 95% CI 0.88, 1.19).
Conclusion: Although black women had higher rates of intrapartum hemorrhage than whites, the increased frequency was attributable to differences in clinical presentation and other risk factors.
Intrapartum hemorrhage, comprising the clinical processes of placental abruption, placenta previa, and other unspecified uterine hemorrhage, occurs in up to 2% of all pregnancies.1 Although infrequent, these complications are associated with increased frequency of neonatal, perinatal, and maternal morbidity and mortality and represent a disproportionately large proportion of the attributable risk for those adverse outcomes.2 Prior studies identified maternal age, hypertension, preeclampsia, smoking, substance abuse, parity, and prior occurrence of hemorrhage as risk factors for intrapartum hemorrhage.3–7 Population surveillance studies have identified an increased frequency of intrapartum hemorrhage among black women,8,9 and several investigators have accordingly adjusted for race in their evaluation of other risk factors.6,10–16 However, it is not known whether race is a separate risk factor or is the aggregate measure of other unmeasured covariates—in short, are racial differences in the frequency of intrapartum hemorrhage real?
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Analyses were limited to live singleton births of infants between 24 and 44 weeks’ gestation occurring in North Carolina hospitals during the study period (n = 840,721). Multiple gestations (n = 21,877), infants delivered before 24 weeks’ gestation (n = 2,761), and infants born after 44 weeks’ gestation (n = 7,001) were excluded. Infants with missing data for placental abruption, placenta previa, or unspecified hemorrhage (n = 757) were excluded. Deliveries with no accompanying clinical data concerning events of pregnancy, labor, or delivery (n = 695) were also excluded. In total, 32,962 (3.9%) births met one or more of the exclusion criteria; all remaining deliveries (n = 807,759) comprised the study cohort.
We first sought to estimate the frequency of any intrapartum hemorrhage complicating a live birth. Intrapartum hemorrhage was defined as a diagnosis of placental abruption, placenta previa, or other unspecified hemorrhage at any point during labor or delivery as noted by three separate check boxes on the birth certificate form. If not checked, the respective form of intrapartum hemorrhage was assumed not to have occurred.
In the second portion of our study, we evaluated the association between maternal race and the occurrence of any intrapartum hemorrhage, placental abruption, placenta previa, and other hemorrhage during labor or delivery. For this analysis, maternal race was categorized as non-Hispanic white, non-Hispanic black, and other minority race. Maternal race was evaluated for its association with the occurrence of any intrapartum hemorrhage, placental abruption, placenta previa, and unspecified hemorrhage by 
2test. In addition, maternal race was also evaluated for bivariate associations with patients’ demographic and clinical characteristics using 2and t tests.
The independent risk of any intrapartum hemorrhage associated with maternal race was then determined by adjusting for maternal demographic and clinical characteristics in multivariable logistic regression analyses. Variables incorporated in the model included demographic characteristics (maternal age, level of education, and marital status), pregnancy history, prenatal care, and complications during pregnancy. Variables evaluated for pregnancy history included maternal gravidity and, among multigravid mothers, prior terminations and interpregnancy intervals. Parity was not incorporated because of high collinearity with gravidity but was tested in place of gravidity in a second model; results were unchanged. Characteristics associated with prenatal care included gestational age of infant, maternal weight gain during pregnancy, use of alcohol or cigarettes during pregnancy, and adequacy of prenatal care as evaluated by the Kessner Index. Finally, clinical complications, including maternal anemia, cardiac abnormalities, acute or chronic lung disorders, gestational diabetes, hemoglobinopathy, renal disease, uterine bleeding during pregnancy, pregnancy-induced hypertension, premature rupture of membranes, and eclampsia, were incorporated. Analyses were repeated to evaluate the specific association of maternal race with placental abruption, placenta previa, and unspecified hemorrhage.
Three secondary analyses were conducted to determine whether any risk of intrapartum hemorrhage attributable to maternal race was limited to particular subsets of high-risk births. In the first cohort, we excluded mothers with evidence of uterine bleeding earlier in pregnancy because those women might have had or been at higher risk for intrapartum hemorrhage, placenta previa in particular. Although there were missing data in less than 3% of all cases, a notable number of birth certificates were excluded from multivariable analysis because of incomplete data for weight gain during pregnancy (n = 22,537; 2.8%). Thus, analyses were repeated after removing pregnancy weight gain from multivariable models to determine whether maternal race remained associated with intrapartum hemorrhage with the addition of previously excluded observations. Finally, to adjust for the possibility that effects might be attributable to a high-risk subset of women with multiple births complicated by hemorrhage, analyses were stratified by year and repeated.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Black women were more likely to be unmarried at time of delivery and were the youngest mothers, whereas women of other minority races had the lowest levels of education. Black women had markedly different pregnancy courses, including lower gestational ages, lower weight gains during pregnancy, and higher gravidity and parity than white women and those of other minority races. Black women also had more frequent occurrences of specific pregnancy comorbidities, including maternal anemia, hemoglobinopathy, acute and chronic lung disorders, eclampsia, and premature rupture of membranes, and more frequent histories of prior terminations. Racial groups were clinically comparable for other pregnancy comorbidities. Finally, black women had the highest rates of inadequate prenatal care and alcohol use during pregnancy, whereas white women had the highest rate of cigarette use during pregnancy.
|
). Black women had the highest rate of any intrapartum hemorrhage (1.52%), although they were statistically comparable to white women (1.47%, P = .12); other minority races had the lowest frequency of any intrapartum hemorrhage (1.33%, P = .006 for three-way comparison). Racial differences in the frequency of intrapartum hemorrhage varied by type of hemorrhage. Black women had higher rates of placental abruption than white women (0.79% versus 0.68%, P = .001), who had higher rates than women of other minority races (0.68% versus 0.56%, P = .002). Race was not associated with the frequency of placenta previa: white (0.43%), black (0.43%), and other minority race (0.40%) women had comparable rates (P = .5). However, white women had the highest proportion of unspecified hemorrhage (0.42%), and black women had the lowest proportion (0.37%), but both groups were comparable to women of other minority races (0.42% versus 0.42%, P = .924 and 0.37% versus 0.42%, P = .07) (Table 2).
|
).
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Although previous studies suggested racial variance in crude rates of placenta previa8 or adjusted for race in their multivariable analysis,6,10–12 we found no such difference. Statistically comparable crude rates of placenta previa might not be surprising given the modest effects of race (OR 1.2) reported by Zhang and coworkers,6,10 the nonsignificant findings of McMahon et al,11 and the trend toward lower risk among nonwhite women reported by Chelmow et al.12 After multivariable adjustment, however, black women were at slightly lower risk of placenta previa (OR 0.90, 95% CI 0.82, 0.98). A modest decrease in risk might be possible given that crude rates of placenta previa between racial groups were comparable, yet black women had significantly more complicated pregnancies, poorer reproductive history, and less adequate prenatal care. Alternatively, a lower risk among black women might reflect geographic variation in placenta previa frequency. As Iyasu et al8 reported, white women in the South have the highest rates of placenta previa nationwide, but black women in the South have the lowest. Regional factors, including socioeconomic characteristics particular to southern whites and blacks, might account for this variation. Furthermore, variations in reporting of placenta previa among black women suggest that estimates of placenta previa frequency might be inaccurate. As such the diminished risk of placenta previa we report might be an artifact of racial variations in case identification rather than a race-associated effect.
Placental abruption is the most common form of intrapartum hemorrhage, occurring in 0.70% of live births and accounting for nearly half of all cases of intrapartum hemorrhage. The frequency we found is markedly lower than that reported by Misra and Ananth15 (1.7% first pregnancy, 2.2% second pregnancy), Ananth et al13 (1.0%), Saftlas et al9 (1.2%), and Omu and Diejomash16 (1.0%). Pritchard et al,14 however, reported a frequency of less than 28 cases per 10,000 births. Variations in frequency are likely attributable to differences in populations and case definitions. The lower frequency reported by Pritchard et al would be expected because their case definition was based on severe placental abruption resulting in fetal death; hence, less severe cases of abruption were not identified. Similarly, some data13,15,16 were based on studies at single or multiple tertiary care centers and thus reflect frequency among a higher risk set of the general population. Conversely, our rates are lower than those reported by Saftlas et al9 in a national evaluation of hospital data. The higher frequency in their cohort, however, might reflect their inclusion of twin gestations, a group at higher risk for placental abruption.
Our multivariable analyses indicated that black and other minority race women were at lower risk of placental abruption than white women. A lower adjusted risk for placental abruption might be plausible. An evaluation of national data concerning placental abruption found relatively comparable rates of placental abruption for black and white women for most the 1980s, except in the mid 1980s.9 Assuming hospital discharge data are of sufficient quality for surveillance purposes and racial variations in the general population are minimal, a lower risk for placental abruption among black women would be expected given the higher rates of comorbidities and poorer prenatal care among black women. Alternatively, the finding of Misra and Ananth15 that black women are at lower risk than white women for placental abruption in a second pregnancy suggests that a lower adjusted risk might reflect differences associated with gravidity. Because most births in both white and black women were to multigravidas, lower risk of placental abruption among black mothers would be expected. Although we adjusted for gravidity in our analyses, unmeasured fertility effects associated with gravidity could explain differences, but the specific mechanism of such a risk is not known.
Racial variation in the frequency of intrapartum hemorrhage is difficult to assess given the absence of data on that outcome. Although other minority race women (OR 1.02, 95% CI 0.88, 1.19) were comparable to white women for risk of unspecified hemorrhage after multivariable adjustment, black women were at lower risk (OR 0.84, 95% CI 0.76, 0.92). This finding is difficult to explain, particularly the possibility of a race-specific effect that would make black race protective for the occurrence of unspecified hemorrhage. The decreased risk suggests racial variation in the identification of unspecified hemorrhage such that unspecified hemorrhages in black women were noted and documented less frequently or differently than those in white women. Alternatively, the definition of unspecified hemorrhage by exclusion of other hemorrhage etiologies is imprecise and might incorporate hemorrhage attributable to other etiologies that were documented erroneously as unspecified. Racial variation in such a process could also explain the adjusted effect reported. Further research is needed to evaluate the manner in which such unspecified hemorrhages are identified. In the interim, results should be evaluated with caution.
Our study has several limitations. North Carolina birth certificates have been validated,17 but they might underreport the frequency of intrapartum hemorrhage through incomplete reporting or might have documented previously resolved intrapartum hemorrhage as an event of labor, thus decreasing the specificity of the diagnosis of intrapartum hemorrhage. In addition, without histologic, ultrasound, or other clinical confirmation, birth certificates might classify the etiology of intrapartum hemorrhage inaccurately. This is of particular concern for unspecified hemorrhages, which are defined by exclusion of alternative hemorrhage etiologies. Inappropriate or overlooked exclusion etiologies, particularly for conditions not listed on the birth certificate form, can be problematic. Similarly, risk factors incorporated in multivariable modeling could be reported inaccurately, particularly for historically poorly reported measures, such as substance abuse.18 Second, although we evaluated intrapartum hemorrhage, hemorrhages might have first manifested antepartum. Thus, our outcome could contain antepartum hemorrhages misclassified as intrapartum events. We were unable to adjust for the effect of prior hemorrhage, which is not only a predictor of subsequent hemorrhage but appears to vary as a risk factor between racial groups.15 Given the large sample size, some effects might not be clinically significant despite being statistically significant. Finally, our inability to adjust for other covariates associated with intrapartum hemorrhage, including cocaine abuse during pregnancy,19 placental dysfunction,20 and socioeconomic status, suggest the possibility of unmeasured residual confounding.
|
Footnotes |
|---|
Received June 8, 2000. Received in revised form September 27, 2000. Accepted October 12, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Atrash HK, Koonin LM, Lawson HW, Franks AL, Smith JC. Maternal morbidity in the United States, 1979–1986. Obstet Gynecol 1990;76:1055–60.
3. Williams MA, Liberman E, Mittendorf R, Monson OR, Schoenbaum SC. Risk factors for abruptio placentae. Am J Epidemiol 1991;134:965–72.
4. Krohn M, Voigt L, McKnight B, Daling JR, Starzyk P, Benedetti TJ. Correlates of placental abruption. Br J Obstet Gynaecol 1987;94: 333–40.[Medline]
5. Ananth CV, Savitz DA, Luther ER. Maternal cigarette smoking as a risk factor for placental abruption, placenta previa, and uterine bleeding in pregnancy. Am J Epidemiol 1996;144:881–9.
6. Zhang J, Savitz DA. Maternal age and placenta previa: A population-based, case-control study. Am J Obstet Gynecol 1993;168: 641–5.[Medline]
7. Raymond EG, Mills JL. Placental abruption: Maternal risk factors and associated fetal conditions. Acta Obstet Gynecol Scand 1993; 72:633–9.[Medline]
8. Iyasu S, Saftlas AK, Rowley DL, Koonin LM, Lawson HW, Atrash HK. The epidemiology of placenta previa in the United States, 1979 through 1987. Am J Obstet Gynecol 1993;168:424–9.
9. Saftlas AF, Olson DR, Atrash HK, Rochat R, Rowley D. National trends in the incidence of abruptio placenta, 1979–1987. Obstet Gynecol 1991;78:1081–6.
10. Zhang J, Fried DB. Relationship of maternal smoking during pregnancy to placenta previa. Am J Prev Med 1992;8:278–82.[Medline]
11. McMahon MJ, Li R, Schenck AP, Olshan AF, Royce RA. Previous cesarean birth: A risk factor for placenta previa? J Reprod Med 1997;42:409–12.[Medline]
12. Chelmow D, Andrew DE, Baker ER. Maternal cigarette smoking and placenta previa. Obstet Gynecol 1996;87:703–6.[Abstract]
13. Ananth CV, Berkowitz GS, Savitz DA, Lapinski RH. Placental abruption and adverse perinatal outcomes. JAMA 1999;282:1646–51.
14. Pritchard JA, Cunningham FG, Pritchard SA, Mason RA. On reducing the frequency of severe abruptio placentae. Am J Obstet Gynecol 1991;165:1345–51.[Medline]
15. Misra DP, Ananth CV. Risk factor profiles of placental abruption in first and second pregnancies: Heterogeneous etiologies. J Clin Epidemiol 1999;52:453–61.[Medline]
16. Omu AE, Diejomaoh FME. Racial differences in the etiology of abruptio placentae. Int J Gynecol Obstet 1981;19:205–10.[Medline]
17. Buescher PA, Taylor KP, Davis MH, Bowling JM. The quality of the new birth certificate data: A validation study in North Carolina. Am J Public Health 1993;83:1163–5.
18. Kline J, Ng SK, Schittini M, Levin B, Susser M. Cocaine use during pregnancy. Am J Public Health 1997;87:352–8.
19. Macones GA, Sehdev HM, Parry S, Morgan MA, Berlin JA. The association between maternal cocaine use and placenta previa. Am J Obstet Gynecol 1997;177:1097–100.[Medline]
20. Rasmussen S, Irgens LM, Dalaker K. A history of placental dysfunction and risk of placental abruption. Paediatr Perinat Epidemiol 1999;13:9–21.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. J. Tucker, C. J. Berg, W. M. Callaghan, and J. Hsia The Black-White Disparity in Pregnancy-Related Mortality From 5 Conditions: Differences in Prevalence and Case-Fatality Rates Am J Public Health, February 1, 2007; 97(2): 247 - 251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.-H. Hung, C.-C. Hsieh, J.-J. Hsu, L.-M. Lo, T.-H. Chiu, and T.-T. Hsieh Risk Factors for Placental Abruption in an Asian Population Reproductive Sciences, January 1, 2007; 14(1): 59 - 65. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
